An approach for revising a fuzzy logic controller using Q-learning algorithm

The use of renewable energies along with the unexpected changes in load and the intermittence in power transmission lines may cause a voltage drop in a stand-alone system and challenge the reliability of the system. One way to indemnify the changing nature of renewable energies in the short term without the need to disconnect loads or turn on other plants is the energy storage. This paper proposes an improvement of a fuzzy logic controller adopted to manage the energy in a stand-alone renewable energy system using the Q-learning algorithm. All the fuzzy rules generated by the controller were based on expert knowledge which means that they are obtained on the basis of observations by human operators. Modified fuzzy rules using the Q-learning algorithm are proposed to incorporate the reinforcement and therefore, present an optimal method for charge and discharge the storage devices which are the batteries.